This paper examines the idea of an extended unit of action, which is the idea that the reasons for or against an individual action can depend on the qualities of a larger pattern of action of which it is a part. One concept of joint action is that the unit of action can be extended in this sense. But the idea of an extended unit of action is surprisingly minimal in its commitments. The paper argues for this conclusion by examining uses of the idea of an extended unit of action in four theoretical contexts. It also explains why the idea of an extended unit of action need not involve magical thinking, and discusses possible replies to an objection based on a worry about recklessness.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
Tax calculation will be finalised during checkout.
Sugden (2000: 191) disagrees. I do not mean to deny that agents sometimes act for reasons that are not genuine normative reasons. The suggestion instead is that a form of explanation of action—for example, in terms of the idea of an extended unit of action—makes sense only if the normative ideas used in the explanation are intelligible. Admittedly, the threshold of intelligibility may be rather low. For empirical evidence that people use ‘team reasoning’, see Colman et al. (2008).
In Woodard (2008b: 63) I defined the idea with respect to ‘the deliberative stance’. For present purposes, the definition I have given in the text is more appropriate (though the two definitions, properly understood, express the same idea).
An action is ‘genuinely joint’ only if it is, or would be, performed by a genuinely extended agent. This claim is deliberately schematic, doing no more than connecting the two concepts of joint action and extended agency; in particular, it does not tell us what is required for either genuine joint action or for genuinely extended agency. I use these terms as placeholders to help characterize the idea of the mirroring assumption. This assumption is general: it can be combined with many different (and rival) views about the necessary and sufficient conditions for the existence of extended agency or of joint action.
I learned the concept of what I call the ‘unit of action’ from Susan Hurley. She tended to use the phrase ‘unit of agency’ instead to describe this concept (1989: 145–148). This usage has been somewhat influential. However, I prefer ‘unit of action’ because it makes it easier to see that the mirroring assumption is a substantive thesis. In later work (e.g. 2005), Hurley used the phrase ‘unit of activity’.
Note that this is not endemic to all theories of rationality that conform to the standard view, since there could be some different basis for the prediction of the other agent’s behaviour, such as salience (Schelling 1980: Ch. 3).
Of course there is an important difference between expected and actual consequences. But since this issue is orthogonal to those discussed in this paper, I will ignore it.
We need not ask why each player values these outcomes in these ways. It could be a matter of self-interest, altruism, theology, or something else. All that is necessary is that the outcomes are valued in the ways shown, and depend on the players’ actions in the ways shown.
Howard (1988) discusses a case in which each player can detect the other’s decision rule, and argues that it would be rational to choose C if each player has a rule which says roughly ‘choose C if and only if the other has this same rule’. He also shows that this need not involve any regress.
X is weakly pareto superior to Y when some prefer X and none prefer Y. X is strongly pareto superior to Y when all prefer X to Y.
The same is true of the concept of Nash equilibrium, and David Lewis’s concept of a ‘coordination equilibrium’ (1986: 14).
‘<C, C>’ here denotes the outcome produced by both players’ choice of C.
Sugden writes that ‘[n]othing in this account of team agency purports to tell people when they ought—whether morally or rationally—to act as members of teams’ (2000: 195, emphasis in the original). One might interpret this as claiming that the theory of Team Thinking has no normative content at all. However, I think that is a mistake. The burden of Sugden’s claim seems to be, instead, that it is a mistake to think that rationality (or morality) requires that individuals think in team-directed ways. But the theory of Team Thinking remains (conditionally) normative in the following sense: it tells one what it is rational to do, if one has the appropriate team-directed attitudes.
I do not claim that understanding possibilism as a form of Plan Consequentialism resolves all problems with it. There are still difficult issues about specifying what the actor could do, for example. Instead, the claim is that understanding it in this way explains why the possible outcomes that possibilists believe are relevant only ever include ones accessible by sequences of action that the actor could perform. This feature of possibilism is otherwise mysterious.
There are limiting cases which complicate this story. Presumably, the duty forbids committing suicide. I shall set these complications aside.
If Parfit (2011) is right that we can think of Kantian contractualism as converging with Rule Consequentialism, and I am right that Rule Consequentialism depends centrally on the idea of an extended unit of action, we might conceive Kant’s ethics as using this same idea more generally, not just with respect to imperfect duties.
The phrase—though not the explanation—is Levi’s (1997: ix).
One complication is that an individual action’s causal properties might be crucial in making it part of the favoured pattern. For example, an act of voting is part of the pattern electing the government only because it has certain causal properties. I am grateful to an anonymous referee for this point.
Recognition of decision procedures is a central feature of Howard’s treatment of one-shot PDs, too (1988).
A variant of this first response to the worry about recklessness claims that pattern-based reasons exist only when all the agents involved together satisfy conditions of joint agency—which presumably include, but go beyond, the sort of cooperativeness that does the work in the previous proposal (Sugden 1993: 87). This is one way of motivating the mirroring assumption described in §1, and it may explain why some claim that the unit of action can be extended only intrapersonally. However, once we’ve seen the possibility of restricting pattern-based reasons to cooperative contexts, it’s not clear that considerations of recklessness speak in favour of further restriction to situations in which there is a prospect of joint agency. This further restriction just seems gratuitous. It’s not clear, for example, that others’ intentions to engage in a joint action with me further reduces the risk of my performing my part in a favoured pattern, as compared with their intentions simply to cooperate with me in producing the best pattern.
One debt this response incurs is to give some account of how act-based and pattern-based reasons may interact in a single case. For discussion see (2008b: 107–118).
If the mirroring assumption is correct, then the idea of an extended unit of action depends on a richer conception of joint action, according to which the larger pattern must be performable by an extended agent. However, I pointed out in §1 that the mirroring assumption is contestable.
Bacharach, M. 1999. Interactive team thinking: A contribution to the theory of co-operation. Research in Economics 53: 117–147.
Bacharach, M. 2006. In Beyond individual choice. Teams and frames in game theory, ed. N. Gold and R. Sugden. Princeton: Princeton University Press.
Bergström, L. 1966. The alternatives and consequences of actions. Stockholm: Almqvist and Wiskell.
Colman, A.M. 2003. Cooperation, psychological game theory, and limitations of rationality in social interaction. The Behavioral and Brain Sciences 26: 139–153.
Colman, A.M., B.D. Pulford, and J. Rose. 2008. Collective rationality in interactive decisions: Evidence for Team Thinking. Acta Psychologica 128: 387–397.
Cummiskey, D. 1996. Kantian consequentialism. New York: Oxford University Press.
Elster, J. 1989. The cement of society. Cambridge: Cambridge University Press.
Feldman, F. 1997. World utilitarianism. In Utilitarianism, hedonism, and desert, ed. F. Feldman, 20–35. Cambridge: Cambridge University Press.
Gold, N., and R. Sugden. 2007. Theories of team agency. In Rationality and commitment, ed. F. Peter and H. Schmid, 280–312. Oxford: Oxford University Press.
Goldman, H.S. 1976. Dated rightness and moral imperfection. The Philosophical Review 85: 449–487.
Hooker, B. 2000. Ideal code, real world. Oxford: Clarendon.
Howard, J.V. 1988. Cooperation in the Prisoner’s Dilemma. Theory and Decision 24: 203–213.
Hurley, S.L. 1989. Natural reasons. New York: Oxford University Press.
Hurley, S.L. 1991. Newcomb’s Problem, Prisoners’ Dilemma, and collective action. Synthese 86: 173–196.
Hurley, S.L. 1994. A new take from Nozick on Newcomb’s Problem and Prisoners’ Dilemma. Analysis 54: 65–72.
Hurley, S.L. 2005. Social heuristics that make us smarter. Philosophical Psychology 18: 585–612.
Jackson, F. 1987. Group morality. In Metaphysics and morality. Essays in Honour of J. J. C. Smart, ed. P. Pettit, R. Sylvan, and J. Norman, 91–110. Oxford: Basil Blackwell.
Jackson, F., and R. Pargetter. 1986. Oughts, options, and actualism. The Philosophical Review 95: 233–255.
Levi, I. 1997. The covenant of reason. Cambridge: Cambridge University Press.
Lewis, D. 1986. Convention. Oxford: Basil Blackwell. originally published 1969.
McClennen, E.F. 1985. Prisoner’s Dilemma and resolute choice. In Paradoxes of rationality and cooperation, ed. R. Campbell and L. Sowden, 94–104. Vancouver: The University of British Columbia Press.
McClennen, E.F. 1990. Rationality and dynamic choice. Cambridge: Cambridge University Press.
Mulgan, T. 2001. The demands of consequentialism. Oxford: Clarendon.
Parfit, D. 1987. Reasons and Persons. Reprint with corrections. Oxford: Clarendon.
Parfit, D. 2011. On what matters. Oxford: Oxford University Press. forthcoming.
Regan, D. 1980. Utilitarianism and co-operation. Oxford: Clarendon.
Schelling, T.C. 1980. The strategy of conflict. Cambridge: Harvard University Press. originally published 1960.
Sobel, J.H. 1976. Utilitarianism and past and future mistakes. Noûs 10: 195–219.
Sugden, R. 1993. Thinking as a team: Towards an explanation of nonselfish behavior. Social Philosophy and Policy 10: 69–89.
Sugden, R. 2000. Team preferences. Economics and Philosophy 16: 175–204.
Woodard, C. 2003. Group-based reasons for action. Ethical Theory and Moral Practice 6: 215–229.
Woodard, C. 2008a. A new argument against rule consequentialism. Ethical Theory and Moral Practice 11: 247–61.
Woodard, C. 2008b. Reasons, patterns, and cooperation. New York: Routledge.
Woodard, C. 2009. What’s wrong with possibilism. Analysis 69: 219–226.
Zimmerman, M.J. 1996. The concept of moral obligation. Cambridge: Cambridge University Press.
I am very grateful to the late Susan Hurley, the late Michael Bacharach, and Alex Gregory for very helpful discussion of these issues. I am especially grateful to three anonymous referees and to Stephen Butterfill and Natalie Sebanz, all of whose extremely useful and generous comments helped me greatly improve this paper.
About this article
Cite this article
Woodard, C. Rationality and the Unit of Action. Rev.Phil.Psych. 2, 261–277 (2011). https://doi.org/10.1007/s13164-011-0058-z
- Joint Action
- Normative Reason
- Standard View
- Extended Unit
- Collective Action Problem